From a6962716ca9c66c24dd6b85d01fadd18c305cc5d Mon Sep 17 00:00:00 2001
From: zhanghongji <1029550448@qq.com>
Date: Thu, 29 Oct 2020 10:36:14 +0800
Subject: [PATCH 1/2] =?UTF-8?q?=E6=A0=B9=E6=8D=AE=E6=9C=80=E6=96=B0?=
 =?UTF-8?q?=E5=8F=8D=E9=A6=88=E5=BB=BA=E8=AE=AE=E6=96=B0=E5=A2=9E=E4=BA=86?=
 =?UTF-8?q?=E4=BB=A5=E4=B8=8B=E4=BF=AE=E6=94=B9=EF=BC=9A=201.=E6=B7=BB?=
 =?UTF-8?q?=E5=8A=A0=E6=95=B0=E6=8D=AE=E9=9B=86=E5=8E=8B=E7=BC=A9=E5=8C=85?=
 =?UTF-8?q?=E4=B8=8B=E8=BD=BD=E9=93=BE=E6=8E=A5=E5=92=8C=E4=BD=BF=E7=94=A8?=
 =?UTF-8?q?=E6=96=B9=E6=B3=95=202.=E6=B5=8B=E8=AF=95pillow=E5=9C=A8paddlep?=
 =?UTF-8?q?addle=E5=AE=89=E8=A3=85=E5=90=8E=E6=98=AF=E5=90=A6=E9=9C=80?=
 =?UTF-8?q?=E8=A6=81=E5=8D=95=E7=8B=AC=E5=AE=89=E8=A3=85=203.=E5=9C=A8?=
 =?UTF-8?q?=E7=9B=AE=E5=BD=95=E4=B8=8B=E6=B7=BB=E5=8A=A0=E6=B5=8B=E8=AF=95?=
 =?UTF-8?q?=E6=95=B0=E6=8D=AE=204.=E6=9B=B4=E6=8D=A2=E7=AC=AC=E4=B8=89?=
 =?UTF-8?q?=E6=96=B9=E8=A7=A3=E7=A0=81=E5=99=A8=E5=B9=B6=E5=AE=9E=E7=8E=B0?=
 =?UTF-8?q?=E9=9D=92=E6=98=A5=E7=89=88=EF=BC=8C=E5=BE=852.0=E6=9B=B4?=
 =?UTF-8?q?=E6=96=B0ctc-decode=E5=90=8E=E5=86=8D=E6=9B=B4=E6=96=B0?=
 =?UTF-8?q?=E8=AF=A5=E5=A4=84=E4=BB=A3=E7=A0=81=E3=80=82?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 paddle2.0_docs/image_ocr/OCR.ipynb           |  44 +++++++++++++------
 paddle2.0_docs/image_ocr/sample_img/9450.jpg | Bin 0 -> 1640 bytes
 paddle2.0_docs/image_ocr/sample_img/9451.jpg | Bin 0 -> 1623 bytes
 paddle2.0_docs/image_ocr/sample_img/9452.jpg | Bin 0 -> 1683 bytes
 4 files changed, 30 insertions(+), 14 deletions(-)
 create mode 100644 paddle2.0_docs/image_ocr/sample_img/9450.jpg
 create mode 100644 paddle2.0_docs/image_ocr/sample_img/9451.jpg
 create mode 100644 paddle2.0_docs/image_ocr/sample_img/9452.jpg

diff --git a/paddle2.0_docs/image_ocr/OCR.ipynb b/paddle2.0_docs/image_ocr/OCR.ipynb
index 370ce582..d46a2db4 100644
--- a/paddle2.0_docs/image_ocr/OCR.ipynb
+++ b/paddle2.0_docs/image_ocr/OCR.ipynb
@@ -22,7 +22,9 @@
     "**数据展示**\n",
     "<p align=\"center\">\n",
     "<img src=\"https://ai-studio-static-online.cdn.bcebos.com/865dd55861e24cfaa601d9f87655776c1458d099b487449f946da9b3138fc700\" width=\"400\"><br/>\n",
-    "</p>"
+    "</p>  \n",
+    "\n",
+    "点此[快速获取本节数据集](https://aistudio.baidu.com/aistudio/datasetdetail/57285)，待数据集下载完毕后可使用`!unzip OCR_Dataset.zip -d data/`命令或熟悉的解压软件进行解压，待数据准备工作完成后修改本文“训练准备”中的`DATA_PATH = 解压后数据集路径`。"
    ],
    "cell_type": "markdown",
    "metadata": {}
@@ -126,16 +128,13 @@
     "\n",
     "CTC相关论文：[Connectionist Temporal Classification: Labelling Unsegmented Sequence Data with Recurrent Neu](http://people.idsia.ch/~santiago/papers/icml2006.pdf) \n",
     "<p align=\"center\">\n",
-    "<img src=\"https://ai-studio-static-online.cdn.bcebos.com/50cf1fc38f6b40e596acf71dc43333ff49dcaafb5a9f484b8aeee2db2c08ca67\" width=\"800\"><br/>\n",
+    "<img src=\"https://ai-studio-static-online.cdn.bcebos.com/f9458cedbb4441d682f15fefd3f3cae5e49d499bcf0a4bbdb976dfdff5a2e656\" width=\"800\"><br/>\n",
     "</p>\n",
     "\n",
-    "网络部分，因本篇采用数据集较为简单且图像尺寸较小并不适合较深层次网络。若在对尺寸较大的图像进行模型构建，可以考虑使用更深层次网络/注意力机制来完成。当然也可以通过目标检测形式先检出文本位置，然后进行OCR部分模型构建。\n",
+    "网络部分，因本篇采用数据集较为简单且图像尺寸较小并不适合较深层次网络。若在对尺寸较大的图像进行模型构建，可以考虑使用更深层次网络/注意力机制来完成。当然也可以通过目标检测形式先检出文本位置，然后进行OCR部分模型构建。(下方样例来源[PaddleOCR](v))\n",
     "\n",
     "<p align=\"center\">\n",
-    "<img src=\"https://ai-studio-static-online.cdn.bcebos.com/6e0665ddfe6a46e1b658da870cd4043f1d50e5f4dc2746018c710c58d2e0c18c\" width=\"400\"><br/>\n",
-    "  \n",
-    "  \n",
-    "<a href=\"https://github.com/PaddlePaddle/PaddleOCR\">PaddleOCR效果图</a>\n",
+    "<img src=\"https://ai-studio-static-online.cdn.bcebos.com/6e0665ddfe6a46e1b658da870cd4043f1d50e5f4dc2746018c710c58d2e0c18c\" width=\"400\"></br>\n",
     "</p>"
    ]
   },
@@ -215,6 +214,8 @@
     "        if self.is_infer:\n",
     "            # 输出层 - Shape = (Batch Size, Max label len, Prob) \n",
     "            x = paddle.nn.functional.softmax(x)\n",
+    "            # 转换为标签\n",
+    "            x = paddle.tensor.argmax(x, axis=-1)\n",
     "        return x"
    ]
   },
@@ -452,7 +453,7 @@
    },
    "outputs": [],
    "source": [
-    "# 待预测目录\n",
+    "# 待预测目录 - 可在测试数据集中挑出\b3张图像放在该目录中进行推理\n",
     "INFER_DATA_PATH = \"./sample_img\"\n",
     "# 训练后存档点路径 - 10代表使用第10个存档点\n",
     "CHECKPOINT_PATH = \"./output/10\"\n",
@@ -505,7 +506,7 @@
   {
    "source": [
     "## 开始预测\n",
-    "> 飞桨2.0 CTC Decoder 相关API正在迁移中，本节暂时使用[第三方解码器](https://github.com/awni/speech/blob/072bcf9ff510d814fbfcaad43b2883ecf8f60806/speech/models/ctc_decoder.py)进行解码。"
+    "> 飞桨2.0 CTC Decoder 相关API正在迁移中，本节暂时使用简易版解码器。"
    ],
    "cell_type": "markdown",
    "metadata": {
@@ -533,7 +534,22 @@
     }
    ],
    "source": [
-    "from ctc import decode\n",
+    "# 编写简易版解码器\n",
+    "def ctc_decode(text, blank=10):\n",
+    "    \"\"\"\n",
+    "    简易CTC解码器\n",
+    "    :param text: 待解码数据\n",
+    "    :param blank: 分隔符索引值\n",
+    "    :return: 解码后数据\n",
+    "    \"\"\"\n",
+    "    result = []\n",
+    "    cache_idx = -1\n",
+    "    for char in text:\n",
+    "        if char != blank and char != cache_idx:\n",
+    "            result.append(char)\n",
+    "        cache_idx = char\n",
+    "    return result\n",
+    "\n",
     "\n",
     "# 实例化预测模型\n",
     "model = paddle.Model(Net(is_infer=True), inputs=input_define)\n",
@@ -547,10 +563,10 @@
     "img_names = infer_reader.get_names()\n",
     "results = model.predict(infer_reader, batch_size=BATCH_SIZE)\n",
     "index = 0\n",
-    "for result in results[0]:\n",
-    "    for prob in result:\n",
-    "        out, _ = decode(prob, blank=10)\n",
-    "        print(f\"文件名：{img_names[index]}，预测结果为：{out}\")\n",
+    "for text_batch in results[0]:\n",
+    "    for prob in text_batch:\n",
+    "        out = ctc_decode(prob, blank=10)\n",
+    "        print(f\"文件名：{img_names[index]}，推理结果为：{out}\")\n",
     "        index += 1"
    ]
   }
diff --git a/paddle2.0_docs/image_ocr/sample_img/9450.jpg b/paddle2.0_docs/image_ocr/sample_img/9450.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..028273bef0376d69a087790cfe3ca30b8fa45fa2
GIT binary patch
literal 1640
zcmbW$do<K(7zgloE@m(z*J{|&(1^rpjd2-aG-w)U<+zoiP|EE{C`^*YiezcXv~(zJ
z(tVN3R4!2oCz@$<8?}vLgvcf4m&^Qi+V<?8(_g#$eb4*;^FHT2@8`VF3-`lsftu?M
z)((I`0DzDW01JT=P{d%c7zITv7OSMBsEpTA!Q*gvZ4FITEnOYb23;K@QP054NN=Ml
znMmAhZEX6TIh9H!8PV)*D0XHRRLbW=5K2l)cpRRfqC%kP6ZI+oIdBV5R|Fft4isV|
zK&m59>InETAj$8kfcP`uuON^pv;qdJsHBXO7u2c&Bm#v(qEQM8XtcaKNxl!z>IxeA
z=G!rv9zocR(OQ%gPN^c9QT1r8Cx6Dk;$TdwlJdIs1Z|y7hDMu>w@@vutZitvPR=en
zm^)dnUfw>t_xSGJcPRMqk&w`^@Yv%g;^Gt7C)0jRKa-JpHY+dxLIL;U&xJ*0<-c65
zxK?@nMs>}-+PeA%-u>3L$4{QNKkMiey!`D|Z(skwplDb;GAbE+H$E{tH~(>AQ3`!p
z`s_jg)EAa~{Q~>Sr7m|N(P$JJ``Lv+#>*#49j%~mj?viefenh*+(=1L)M9W-s~#zl
zEj(w|9*p5DuQQ;EHqCyf{VDr*u+;x0`y2L;YY^a32>Iio)Bzp5TQL^7uknYrIhQ1%
zT)AOXNOvUBW}q-ToMn*1vY+&ZLRx!0+qW)a@^QvP)#nH;3|n2{X|Iw3KgM#%iiwK#
zGF?Vbt}0umyDyaN6!zsbOg>>JHI*K&Oo)lJO=Q1(8j(myl^lt3*n6tw3OnP%mS(eF
zJv@>&UV48I^QI88Xq}8A$XfWL;1~=JW=`woM0d2GUNBnGf3qG2rBO^6v=Ex1q5`V0
zPB2x*6?Zj-zS!^Y_{7P|A<hRNS<gC9+C70Lnr?zozFjak&Z~2=+V4}12}jCCIW$|*
zP<E@~#G)|YZqMj2KPNw#+6NpAUeuSv;4MC2#f>zRkq{tDZzRD09E8DexT?-X{`dh>
zNZk=u$a#i=S85?Tj}Dl1v4htqkg6(BhLfA^>!7Unm%_wODSDmhMt63a?mAggD5>>m
zxa#3j8+jWl(txbnE-~{(&oz(H+WMKP_BO;M_L6tN%$au@3p0}ZqvNZ>&HG97{$kmu
zN$BOnLzb=Yh6mpWrpKmKtG#RSq^?Is5@pA+&=F!PvtevSHB8j8u$EVy$ggzKjxrRb
z;LalgeH`hD#MNkD-tthz(9Kxva8Cx+&7!rgCMoY}`-usoOV`FLt4W-0)ozCa7JZdk
zJ}}6E!5n!Yxt~S<=zq^E!b|)1u4|4F6;Mz`F*S|7noFqHySv0!#P#JiPPX4>*FNDE
z+(^I>I;kaQSrX?255j9#38UM<|7f6QaeJwMNTt<r_iP7U!9#k*zHjXJOzp1YY&U%N
zvWvq(UKTHks!Fhe`lAk3Pz+x{yJfk5N-L(A^VD%oN=rxdK)UA<PG0}}+jrNT6Ga%G
zU=#dZB6Zw&^$)u41&7eh4K~DnC|UUF3Y0n+F?>_y@nG@%^}<Z=Y+$>_U707fk&02W
z_08!p7;E!|L9==CY-f>B9F<`<8Y=qrRlI4G{_n)VNEg%B#QAg^Y5T5Pira&svlS!7
zi*vTQI^(euO{Czu-R!{g-;+J^Aha4C>E2}ire=CL?Z7N1>v&f|QSXaOh012m<Sjld
zv9`$%jZC}1z(5$ZN@to^pa_ywvQ<W7u53+~w^PrVDj1xWw~x4>j6=Dmy-iyGHvVL9
zbJ&&D!n@?loyCjns*2I<St)O+gpy!+(Xcq;MnZMhZh<5_lj~kP)GA8Z#3)Pa<W6$>
zPfYM6bWtJ{AUkE5yF4%d1jO>MrO$6Ry7A8U<S;zm`q628;WCTnyv2N!HQ~1Q*(q13
opgf_gjk^?A9GR@ee|>B84%uaYsc!UK)tKq#(&@wB_P_&w06l}`^Z)<=

literal 0
HcmV?d00001

diff --git a/paddle2.0_docs/image_ocr/sample_img/9451.jpg b/paddle2.0_docs/image_ocr/sample_img/9451.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1fbea8ae9dcd92d901b2cd6f0773d620fb42e22a
GIT binary patch
literal 1623
zcmbW#c~Fx_6aerIUm%2pBOG!CL%5YYfLtO545y&n3Wx%7rXZID0p(H@1QhT<8Y`fJ
zT9KQemcz(Ztx!P_gOYM9RY(LSSmbNaPCMgYZQsuB?9A@GeZP4NkHEhH2`4*8JAgm{
zfZz`Rj{;jDEF^>y5)?+EP-wKU2v%AYi@{*!C8flrl@tgYl@#!JWi>rbWfdJ&JYLIm
zla7I*v9U2h)7;98v{lc@n6#b*fkvaT7_6MAs2quiCzAeiz#RZ53^oEg2%!Q5a0m#8
zfI9(!U#B4AOMt%uApju-g;2t15e)xAlLQbzK#%|u5)?!t`L~bq&jAuAC`mLV3rV>J
zqEup~Ng0LZ!m2iP4`ke*a@351;xf@98)W6=71T8}wKi=wHrZloW^Q4-gJNgz;OOMx
z>E-RSXRmMY!I03f@QBFxgu{unq$9~$$4_LRJe8AMRDACIg^MMZN-M5jyIxs!<L0gU
zhQ_95M$6rM%!i#_-95cOKN=W(HZ=Tv<OOSdVsdJFX7=@)Iqv(#r4JvMc`KjRT?hbu
zW%1Xquy0&AzDod!gpjCp7eau>pAZfyNHi3ZB)g#kW2IC`8N$*wh2?b*(5go69GRfF
zry?8FjK|fv>$ESje+SF_U$Vbp|F~EH3nBQ;gK)qayiV00-={FiR8V^|o-al#-o>C@
z2$xE!%XidV4x`V~hs!vX#~GA`RFjI5>CL(9Q2P;^chEUGe=U%F<(LkS?94-vDTP-<
z2i|oY?)F@9?(J2z-c8LNT6$vY#GTKlhN%BS(|Htq@NSs4d2qKMvIqvGI_Qw~f=F({
z=c|!3^-ncu1WtC6-`a_40t`S93?{;1aNLN=Uee`nc#td%q<p>OAmLLDmsEklS&6Y1
zAKpBRaWJc~=0^W`IHiWSljt=VIt1;wA2%R3MUjkNd-F0e<AJ-PW(<a|GWFIYLi;jp
zmH5L37?e}(VbCGh&MUPZ$8s2-F}~JopbG|_L)GyfYrO-Z_jl(jGrRA^$mjG-7PwmF
z2Zs}a@0O0a*v<J}cK2K?)RSp0(=^W`WsD^BzcQO>h;OL#rtqt@-Kq-%o8dDsNH<L5
zn&8JoIjk}R-hSJ}e&X9nw=*|vJ*d8-l{cfAa}%q6cLE#&&z~x8dGy}Z@T_cma912#
ztcsDz_D+eINePT=$qvA}(KCoWkBe9LbPI(2x_fFN*woUIQr(v_F?-UuH+55F!V}D{
zK|Rs;q5IDI&ASwZ&@xf}c&R?4tgOt&V^%QeJskprKrI+7c{cj&&=1mRvWQZ3FSh6H
zy%vdMH4T!lv+8nO(TWmMbUZ*?S^8L3&wU)*T3_2#-_TMJK2L6;Gv%WyEHl(w#YGE@
z7_9nKYi`ZMWS{J%R*Mu}ugb>8T(rDO-p!Eh*YbMBNLJcT?8*Jv)1~t>+Pa<;X#vOc
z6=&2IJdwv1Yx?_Aw3aI9D<;kX(~kqH%6{J>Nwv@!zQD+QRVm^flS%0~;IP?M-T^6*
zq)eZR4WLen8u>SKWwTXVx#FW5r3+C%v4b4e`bH;LYQ+?b#Kf16&h@YAw`0=!v+H1R
zjQ?jQ&!sW{;4oHsRKoI%{+hW5+DS<yF^oF#$|B0yHiY4QBrsNsBCnrLhB#_#a<^m`
z?Q5tCiX)~awTvr<kcx6*E!J*(aARKbb0ZjF7MSeC@!M(al6KxSfz!6iZ|*jQp3f_%
zVNl?$_x*P?`@m{mGyCDEuGZ@xZZla)S-Jz*^wC=K#`FVp$s_pqy<_;n;pML6wpOKv
zuBw;aDklz}JU};a??b$GnpKOx{|vY1+q`byyo=9o-TCs6HlrNXBU8AIFw&#v=;9)6
zZm!qn-=fKu>u*bO^S{{1&bg<bw`zK*Dq@(ybx~sV`A!&nv2f1`+QszRca#Kj_|9c(
G_{E>T6vY()

literal 0
HcmV?d00001

diff --git a/paddle2.0_docs/image_ocr/sample_img/9452.jpg b/paddle2.0_docs/image_ocr/sample_img/9452.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ff4fdd93f12ae33541b035eaa19515527d209b17
GIT binary patch
literal 1683
zcmbW#c{tQ*902h5XT}^%7{=KcVa8ENV;q$s*PxS9>7augwQWT^BRR85(zfHO5T%-=
z?KVS6QZmHMqBRL^7`J3lOlz9oj@oDUdG?<@zVGwCf4uMeyzlq(zF)W<?gknw8NLhv
zK>$FC2H;NM0Wc^O8l{XuqtRF_Mg>Qhhf`I>>1%4K6ATTENQMSP;sUab`2rIwQzFsA
z(bCFx35`ZGGG9t}qB`5y(WrBWKv*mmr;5{?H&2ghOf;tc=YXF8JO+>e0|A);B|L<{
zLvSrHQsk)&eF^YKKuQRtG762ss;DY9a5aDugg_`E5z5L)q+)lxVjUpy%9_SY+)!Hn
zyU-?w2-IZuC5);2-5TxnEfTWb?x<r}6&+nYeFKV_xrOCIn!SUg)6!)go?Z-ZAK#T5
z0yb{i9Jpm`Na&uh@QBF0(TBfdvSN<JrhK1z{6yNx^vta6oZPeL&gT_hzH*gwt>k*?
zz58WcUU|iXhfkl?KCk1~H@s+l`AeJVRr~9Xo?dZZ|G?nT@W{mE)Q6ALGty7L&AA|e
z_{vg@Utxc9;T0|=Bocu{&$%EamZBr@NM++CC`~ti^sYl%Ce&mM!JU2SZVlGdZoNc%
zcT|gt4w=?NnV6$}k^MW^vHvCe6ZV&@1K<#l;`0!A;0oS7GVW2wbx9YwRa8jcv#nZ(
zMdRemEsUmjA?tl9QNHxiaJfd2N@DZfVi<HIcgWUkl#Go}v!1>&n{AwO5fndjXc9Bq
zp}M1oZ>^I@o|RFGrbturmT4G#imVh)P87X=_lX(b#Vkp6v_2quZdfieb|JvvQivA}
zo?xn_d9FQJ$+gd_28-oDF;&}E5yTx=b6eO*XIJrj9ao6sc4cd7r;Q5jGG_Gb2RrO@
zvp8|VvpP8xL4p5@7tKDzlcAlN7YmtU_1t<&O!xI-!l!I*+q^{={^=$WtM)?FsD~u0
zA#98|L%sR%#p&Z28i8#7c1y!Fp<C~L?yh~^Z*`<$bd$FI-PK|DAEeuH_xSny=il10
zDywpw<mN1jP3J3;#;i+q9SUqolqfBho1Kxm$63O_z3vnYlJJQW_Nf!|0}3t63{@y*
z*1;clSJYm0tkw5m_~=9B8Ehd$94LLg*EXV>F!mvc+P*DtWYf06vehQf4ps_3rz|kM
zo5bvP@QALl_Uh;_3EDpJ*h$LrY>m?nKg1dOEk0XNQcTMgHO$wZ>~o^5aH=OB_20Uq
zOjnk+$#q)m3=F(hd2M&F3lHj<4Xu_2T^=5zK0QE-*tyBnxeRf!UWi&tzmWrjygdp;
zk&OQc2KPUw$hUqw^{7caTj{bN1}wu-T~BFS&nhF8$XL4k{B@y-cd*7{(j+v@)6b`=
zAR{9qy~?o;2I|9`C%1X4ajg!m49Q$te0n0~n154#)jC0?GsG((i)XrQSh1XIUta4&
zuc-B{$Ns*KrU}_qKO5(!sAWmEKK)68=yJ`@n8D}4q{pobit6u$*uUxQ98(Fn@srmN
z)F;!(<Z4rA5lco>SG>2;&BV9os-^u#lIFRBx|qO)LBehrh_e~jY<RDN;^WfS^1Ua#
zR~K!NM8_8{B_E9xMX^i1DZ!~78A60-d>HQ9Gcr}A8*}^FRAfTX8(zV{<mR%hBLP;a
zi1Bv(IgUdgyMa$^hQaY<MS&td-kzP(g#od&Sr#SPj&H=p2H&yn({q49_=!eR#STS%
z8ZH-;2kr0NI_%=iJ(~Zs+Nd?L#U_hcuIqOX6`B$|p4D$|%*3TOKs8RB`60cF({1<V
z(QEBq4>MghtaXXFa=u(X<V0usW;RwH(ARgml_=L`C3{u>_>jb5zyDQddBo;@A@h%u
zd#d>MVK2pxlXzJII_HL!zYqJyshAvLH-CT>=9*Gc7d^vR_%LGqg`ATv^dq*?WiJMN
zkbNU~{ykqZ#+79$Y};u!?sX6bganJBLI;U}m(S@K!LQNo6$+va7LH*?lerDTKxT2N
Ku8*Jye*HTbSoTB!

literal 0
HcmV?d00001


From 8e1b29d471ff20b2f62b2098d17e276fdc38058c Mon Sep 17 00:00:00 2001
From: zhanghongji <1029550448@qq.com>
Date: Thu, 29 Oct 2020 11:35:28 +0800
Subject: [PATCH 2/2] =?UTF-8?q?=E9=80=82=E9=85=8D2.0RC0?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 paddle2.0_docs/image_ocr/OCR.ipynb | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/paddle2.0_docs/image_ocr/OCR.ipynb b/paddle2.0_docs/image_ocr/OCR.ipynb
index d46a2db4..db68b55e 100644
--- a/paddle2.0_docs/image_ocr/OCR.ipynb
+++ b/paddle2.0_docs/image_ocr/OCR.ipynb
@@ -163,16 +163,16 @@
     "        self.is_infer = is_infer\n",
     "\n",
     "        # 定义一层3x3卷积+BatchNorm\n",
-    "        self.conv1 = paddle.nn.Conv2d(in_channels=IMAGE_SHAPE_C,\n",
+    "        self.conv1 = paddle.nn.Conv2D(in_channels=IMAGE_SHAPE_C,\n",
     "                                  out_channels=32,\n",
     "                                  kernel_size=3)\n",
-    "        self.bn1 = paddle.nn.BatchNorm2d(32)\n",
+    "        self.bn1 = paddle.nn.BatchNorm2D(32)\n",
     "        # 定义一层步长为2的3x3卷积进行下采样+BatchNorm\n",
-    "        self.conv2 = paddle.nn.Conv2d(in_channels=32,\n",
+    "        self.conv2 = paddle.nn.Conv2D(in_channels=32,\n",
     "                                  out_channels=64,\n",
     "                                  kernel_size=3,\n",
     "                                  stride=2)\n",
-    "        self.bn2 = paddle.nn.BatchNorm2d(64)\n",
+    "        self.bn2 = paddle.nn.BatchNorm2D(64)\n",
     "        # 定义一层1x1卷积压缩通道数，输出通道数设置为比LABEL_MAX_LEN稍大的定值可获取更优效果，当然也可设置为LABEL_MAX_LEN\n",
     "        self.conv3 = paddle.nn.Conv2d(in_channels=64,\n",
     "                                  out_channels=LABEL_MAX_LEN + 4,\n",
@@ -287,8 +287,8 @@
     "        super().__init__()\n",
     "\n",
     "    def forward(self, ipt, label):\n",
-    "        input_lengths = paddle.tensor.fill_constant([BATCH_SIZE, 1], \"int64\", LABEL_MAX_LEN + 4)\n",
-    "        label_lengths = paddle.tensor.fill_constant([BATCH_SIZE, 1], \"int64\", LABEL_MAX_LEN)\n",
+    "        input_lengths = paddle.tensor.creation.fill_constant([BATCH_SIZE, 1], \"int64\", LABEL_MAX_LEN + 4)\n",
+    "        label_lengths = paddle.tensor.creation.fill_constant([BATCH_SIZE, 1], \"int64\", LABEL_MAX_LEN)\n",
     "        # 按文档要求进行转换dim顺序\n",
     "        ipt = paddle.tensor.transpose(ipt, [1, 0, 2])\n",
     "        # 计算loss\n",